def _make_file(self, name, filename, mimetype, body, default_language):
from webpage import WebPage
if type(name) is not str:
raise TypeError, 'expected string, got %s' % repr(name)
# Web Pages are first class citizens
if mimetype == 'text/html':
body = tidy_html(body)
class_id = 'webpage'
elif mimetype == 'application/xhtml+xml':
class_id = 'webpage'
else:
class_id = mimetype
cls = self.database.get_resource_class(class_id)
# Special case: web pages
kw = {'filename': filename, 'data': body}
if issubclass(cls, WebPage):
kk, kk, language = FileName.decode(filename)
if language is None:
text = XHTMLFile(string=body).to_text()
language = guess_language(text) or default_language
kw['data'] = {language: body}
return self.make_resource(name, cls, **kw)
def _make_file(self, name, filename, mimetype, body, default_language):
from webpage import WebPage
if type(name) is not str:
raise TypeError, 'expected string, got %s' % repr(name)
# Web Pages are first class citizens
if mimetype == 'text/html':
body = tidy_html(body)
class_id = 'webpage'
elif mimetype == 'application/xhtml+xml':
class_id = 'webpage'
else:
class_id = mimetype
cls = self.database.get_resource_class(class_id)
# Special case: web pages
kw = {'filename': filename, 'data': body}
if issubclass(cls, WebPage):
kk, kk, language = FileName.decode(filename)
if language is None:
text = XHTMLFile(string=body).to_text()
language = guess_language(text) or default_language
kw['data'] = {language: body}
return self.make_resource(name, cls, **kw)
def extract_archive(self, handler, default_language, filter=None,
postproc=None, update=False):
change_resource = self.database.change_resource
for path_str in handler.get_contents():
# 1. Skip folders
path = Path(path_str)
if path.endswith_slash:
continue
# Skip the owner file (garbage produced by microsoft)
filename = path[-1]
if filename.startswith('~$'):
continue
# 2. Create parent folders if needed
folder = self
for name in path[:-1]:
name, title = process_name(name)
subfolder = folder.get_resource(name, soft=True)
if subfolder is None:
folder = folder.make_resource(name, Folder)
folder.set_value('title', title, default_language)
elif not isinstance(subfolder, Folder):
raise RuntimeError, MSG_NAME_CLASH
else:
folder = subfolder
# 3. Find out the resource name and title, the file mimetype and
# language
mimetype = guess_mimetype(filename, 'application/octet-stream')
name, extension, language = FileName.decode(filename)
name, title = process_name(name)
language = language or default_language
# Keep the filename extension (except in webpages)
if mimetype not in ('application/xhtml+xml', 'text/html'):
name = FileName.encode((name, extension, None))
# 4. The body
body = handler.get_file(path_str)
if filter:
body = filter(path_str, mimetype, body)
if body is None:
continue
# 5. Update or make file
file = folder.get_resource(name, soft=True)
if file:
if update is False:
msg = 'unexpected resource at {path}'
raise RuntimeError, msg.format(path=path_str)
if mimetype == 'text/html':
body = tidy_html(body)
file_handler = file.get_handler(language)
else:
file_handler = file.get_handler()
old_body = file.handler.to_str()
file_handler.load_state_from_string(body)
if postproc:
postproc(file)
# FIXME Comparing the bytes does not work for XML, so we use
# this weak heuristic
if len(old_body) != len(file.handler.to_str()):
change_resource(file)
else:
# Case 1: the resource does not exist
file = folder._make_file(name, filename, mimetype, body,
language)
file.set_value('title', title, language=language)
if postproc:
postproc(file)
def extract_archive(self, handler, default_language, filter=None,
postproc=None, update=False):
change_resource = self.database.change_resource
for path_str in handler.get_contents():
# 1. Skip folders
clean_path = "/".join([
checkid(x) or 'file'
if x else 'file' for x in path_str.split("/")])
path = Path(clean_path)
if path.endswith_slash:
continue
# Skip the owner file (garbage produced by microsoft)
filename = path[-1]
if filename.startswith('~$'):
continue
# 2. Create parent folders if needed
folder = self
for name in path[:-1]:
name, title = process_name(name)
subfolder = folder.get_resource(name, soft=True)
if subfolder is None:
folder = folder.make_resource(name, Folder)
folder.set_value('title', title, default_language)
elif not isinstance(subfolder, Folder):
raise RuntimeError, MSG_NAME_CLASH
else:
folder = subfolder
# 3. Find out the resource name and title, the file mimetype and
# language
mimetype = guess_mimetype(filename, 'application/octet-stream')
name, extension, language = FileName.decode(filename)
name, title = process_name(name)
language = language or default_language
# Keep the filename extension (except in webpages)
if mimetype not in ('application/xhtml+xml', 'text/html'):
name = FileName.encode((name, extension, None))
# 4. The body
body = handler.get_file(path_str)
if filter:
body = filter(path_str, mimetype, body)
if body is None:
continue
# 5. Update or make file
file = folder.get_resource(name, soft=True)
if file:
if update is False:
msg = 'unexpected resource at {path}'
raise RuntimeError, msg.format(path=path_str)
if mimetype == 'text/html':
body = tidy_html(body)
file_handler = file.get_handler(language)
else:
file_handler = file.get_handler()
old_body = file.handler.to_str()
file_handler.load_state_from_string(body)
if postproc:
postproc(file)
# FIXME Comparing the bytes does not work for XML, so we use
# this weak heuristic
if len(old_body) != len(file.handler.to_str()):
change_resource(file)
else:
# Case 1: the resource does not exist
file = folder._make_file(name, filename, mimetype, body,
language)
file.set_value('title', title, language=language)
if postproc:
postproc(file)